Hindawi Publishing CorporationThe Scientific World JournalVolume 2013, Article ID 824979, 6 pageshttp://dx.doi.org/10.1155/2013/824979
Research ArticleCatalytic Oxidation of Propylene, Toluene, Carbon Monoxide,and Carbon Black over Au/CeO2 Solids: Comparing theImpregnation and the Deposition-Precipitation Methods
Antoine Abouka\s,1,2 Samer Aouad,1,2,3 Houda El-Ayadi,1,2 Mira Skaf,1,2,3 Madona Labaki,1,2
Renaud Cousin,1,2 and Edmond Abi-Aad1,2
1 Universite Lille Nord de France, 59000 Lille, France2 Equipe Catalyse, UCEIV, E.A. 4492, MREI 1, ULCO, 145 avenue Maurice Schumann, 59140 Dunkerque, France3 Department of Chemistry, University of Balamand, P.O. Box 100, Tripoli, Lebanon
Correspondence should be addressed to Antoine Aboukaıs; [email protected]
Au/CeO2 solids were prepared by two methods: deposition-precipitation (DP) and impregnation (Imp). The prepared solids werecalcined under air at 400∘C. Both types of catalysts have been tested in the total oxidation of propylene, toluene, carbon monoxide,and carbon black. Au/CeO2-DP solids were the most reactive owing to the high number of gold nanoparticles and Au+ species andthe low concentration of Cl− ions present on its surface compared to those observed in Au/CeO2-Imp solids.
1. Introduction
In recent years, gold-based catalysts receive continuousattention due to their role in various catalytic reactions ofcommercial and environmental importance. Among thesereactions we can quote the oxidation of carbon [1–5] andvolatile organic compounds (VOC) [6–11]. The catalyticactivity and stability of gold-based compounds depend onthe gold particle size, presence of gold cations, metal oxidesupport,method of preparation, calcination temperature, andpretreatment procedure. However, the preparation methodhas been reported to directly affect the reactivity of gold-cerium oxide system. The currently employed methods ofimpregnation, deposition-precipitation, coprecipitation, andchemical vapour deposition strongly influenced the catalyticactivity due to the large differences in gold particles sizeand/or to the availability of active gold sites in close contactwith the support defects on the surface [12–14].
Cerium oxide (CeO2) has received considerable attention
especially in oxidation catalysis [15–20]. This is due to itslow temperature reducibility and its oxygen storage andrelease properties in the presence of noble metal particles.
The oxidation/reduction couple (Ce3+/Ce4+) of ceria particleswhich are in direct contact with the metal particles promotesthe catalytic activity in most cerium-based materials.
The aim of this work is to correlate between the prepa-ration method and the catalytic performance of Au/CeO
2
catalysts in the total oxidation of propylene (C3H6), toluene
(C7H8), carbon monoxide (CO), and carbon black (CB).
2. Materials and Methods
2.1. Preparation of Gold Catalysts. The gold-based catalysts(Au/CeO
2) were prepared using two different methods:
impregnation (Imp) and deposition-precipitation (DP). Theused CeO
2support was prepared according to [16] and was
calcined under air flow (2 L⋅h−1) at 400∘C (1∘C⋅min−1) for 4hours.
2.1.1. The Impregnation Method. One gram of the support(CeO2) was added to 100mL of an aqueous solution of tetra-
chloroauric acid (HAuCl4) containing the suitable amount of
gold.The solution was left under stirring for two hours before
2 The Scientific World Journal
its evaporation at low pressure and 60∘C. The obtained solidwas dried at 80∘C overnight before calcination under air flow(2 L⋅h−1) at 400∘C (1∘C⋅min−1) for 4 h.The obtained solids aredesignated xAu/CeO
2-Impwhere x is the gold weight percent
in the calcined solid.
2.1.2. The Deposition-Precipitation Method. One gram of thesupport (CeO
2) was added to an aqueous solution of tetra-
chloroauric acid (HAuCl4) at 80∘C containing the suitable
amount of gold. The pH of the solution was adjusted to 8by adding NaOH drop by drop under stirring during 4 h.The suspension was filtered and washed several times withhot water in order to eliminate Na+ and Cl− ions. The solidwas then dried in the oven at 100∘C followed by a thermaltreatment under air at 400∘C (1∘C⋅min−1) during 4 h. Theobtained solids are designated xAu/CeO
2-DP where x is the
gold weight percent in the calcined solid.
2.2. Catalytic Activity Measurements. The catalytic oxidationtests of gaseous molecules (CO, C
3H6, and C
7H8) were
carried out at atmospheric pressure in a continuous flowU-shaped reactor with an internal frit (𝜙 = 6mm) onwhich the catalytic bed is placed. 100mg of the powdercatalyst was loaded in the reactor and then placed in anelectrically controlled heating furnace. The reactant gaseswere injected in the system using mass flow controllers forC3H6(6000 ppm) or CO (1000 ppm) and a liquid satura-
tor for C7H8(2000 ppm). The total flow was adjusted to
100NmL⋅min−1 using CO2free industrial air. The product
gases were injected into a VARIAN 4900microgas chro-matography allowing the measurement of C
3H6, CO, C
7H8,
CO2, N2, and O
2in the stream. Prior to any test, the
catalyst (100mg) was reactivated under air flow (2 L⋅h−1)at 400∘C (1∘C⋅min−1) during 1 h. The catalytic test towardsthe combustion of carbon black (CB) (N330 DEGUSSA:specific surface area 𝑆sp = 76m2⋅g−1, elementary analysis:97.23 wt.% C; 0.73wt.% H; 1.16 wt.% O; 0.19 wt.% N; 0.45wt.% S) was studied by simultaneous thermogravimetric(TG)—differential scanning calorimetry (DSC) analysis witha NETZSCH STA 409 apparatus. Before test, 2 wt.% of CBand 98wt.% of catalyst were grinded together in an agatemortar for 15 minutes to ensure a tight contact. 50mg of themixture was then loaded in an alumina crucible and heatedup to 700∘C (5∘C⋅min−1) under air flow of 75mL⋅min−1.Each catalytic test was repeated two times for reproducibilitychecking.
3. Results and Discussion
3.1. Characterization of Catalysts. The solids obtained usingthe two different preparation methods have been separatelycharacterized in a previous work [21, 22].The specific surfaceareas of the impregnated solids (Imp) were smaller comparedto those obtained for the solids prepared by the deposition-precipitation (DP) method. In addition, relatively high con-tents of chloride anions were present in Imp-solids comparedto DP-solids.
Scanning electron microscopy (SEM) and transmissionelectron microscopy (TEM) showed a heterogeneous distri-bution of gold particle sizes at the surface of the Imp-solids.In fact,more than 60%of goldwas present in the formof largeparticles (>100 nm) and about only 20% in the form of smallnanoparticles (<10 nm) in Imp-solids. On the other hand,only nanoparticles with an average diameter estimated to beequal to 3.9 nm have been detected on the DP-solids. Thepresence of large particles in the Imp-solids and its absencein the DP-solids were confirmed using the X-ray diffractiontechnique [21, 22].
The X-ray photoelectron spectroscopy (XPS) showed thepresence of Au0 and Au+ species on both solids types. TheAu+ species represented only 10% of the total gold atomson the Imp-solids surface, while its content was about 20%on the DP-solids surface. The small amount of Au+ thatwas still present after calcination under dry air I attributedto gold species located in the proximity of O2− and/or Cl−ions present in the CeO
2support. Since elementary analysis
showed that the number of Cl present in the DP-solidsis negligible compared to that present in the Imp-solids,therefore the ions surrounding the Au+ species in both solidstypes are more likely to be O2− [21, 22].
With the diffuse reflectance ultraviolet/visible spec-troscopy (DR-UV/Vis), it was demonstrated, according to theplasmon resonance principle, that gold nanoparticles withdiameter smaller than 10 nm are present on both solids types.The intense band obtained for the 4Au/CeO
2-DP catalyst
indicated that the number of nanoparticles in this solid isgreater than that in the impregnated equivalent. In addition,the band obtained at a shorter wavelength demonstratedthat the nanoparticles in the DP-solid have spherical shapesand are in weak interaction with the surface of the ceriasupport [21, 22]. On the contrary, nanoparticles present inthe 4Au/CeO
2-Imp catalyst can be modelled as hemispheres
having more significant interactions with the support surface[21, 22].
The temperature programmed reduction (TPR) resultsshowed that the reduction of Au+ into Au0 occurs at low tem-perature on both 4Au/CeO
2-DP and 4Au/CeO
2-Imp solids
with a reduction peak at 94∘C and 182∘C, respectively. Thepresence of only gold nanoparticles in the first solid facilitatesthe reduction of Au+ species relatively to those present in4Au/CeO
2-Imp. Consequently, the O2− ions situated near the
edge of gold particles in 4Au/CeO2-DP solid are probably
more mobile compared to those in 4Au/CeO2-Imp solid
[21, 22].
3.2. Catalytic Performance of the Solids
3.2.1. Oxidation of Gaseous Probe Molecules. Figure 1 rep-resents the conversion of propylene, toluene, and carbonmonoxide as a function of temperature in the presence ofthe CeO
2support and the gold-based catalysts. Considering
propylene oxidation (Figure 1(a)), adding gold to CeO2leads
to its conversion at lower temperatures with a selectivity of100% towards CO
2formation.
The Scientific World Journal 3
0
20
40
60
80
100
100 150 200 250 300 350 400
Prop
ylen
e con
vers
ion
(%)
CeO2
4Au/CeO2-DP4Au/CeO2-Imp
0.5Au/CeO2-Imp
Temperature (∘C)
(a)
0
20
40
60
80
100
100 150 200 250 300 350 400
Tolu
ene c
onve
rsio
n (%
)
CeO2
4Au/CeO2-DP4Au/CeO2-Imp
0.5Au/CeO2-Imp
Temperature (∘C)
(b)
0
20
40
60
80
100
25 50 75 100 125 150 175 200
CO co
nver
sion
(%)
CeO2
4Au/CeO2-DP4Au/CeO2-Imp
0.5Au/CeO2-Imp
Temperature (∘C)
(c)
Figure 1: Conversion of (a) propylene, (b) toluene, and (c) CO versus temperature over CeO2
, 0.5Au/CeO2
-Imp, 4Au/CeO2
-Imp, and4Au/CeO
2
-DP solids.
The 4Au/CeO2-DP catalyst presents the best activity in
the considered reaction. The 4Au/CeO2-Imp catalyst is less
active which is probably due to its relatively large goldparticles compared to those obtained by the DP method. Tocheck the relation between gold particles size and perfor-mance in oxidation reactions, a catalyst with 0.5 wt.% goldwas prepared using the impregnation method (for low goldcontents smaller gold particles are obtained [22]). This latterismore active than the 4Au/CeO
2-Impbut remains less active
than the 4Au/CeO2-DP catalyst.
Figure 1(b) shows that the impregnated catalyst is lessactive than the bare support in toluene oxidation withT50% equal to 296∘C and 273∘C for 4Au/CeO
2-Imp and
CeO2, respectively. Moreover, up to a conversion of 20%,
CeO2, 0.5Au/CeO
2-Imp, and 4Au/CeO
2-DP exhibit similar
catalytic behaviour. However, CeO2becomes relatively less
active for conversions higher than 20%, while 0.5Au/CeO2-
Imp and 4Au/CeO2-DP solids remain equally active up to
a conversion of 65%. For higher toluene conversions the4Au/CeO
2-DP catalyst becomes the most active among all
the tested solids. It is worth to note that in both propylene andtoluene oxidation reactions, the only products that evolvedwere CO
2and H
2O.
Figure 1(c) shows the evolution of the CO conversion asa function of temperature in the presence of the differentsolids. The support presents a low catalytic activity and theconversion reaches only 6% at 200∘C. The conversion is alsorelatively low for both impregnated catalysts; however, it isslightly higher for 0.5Au/CeO
2-Imp especially when the tem-
perature exceeds 125∘C. On the other hand, the 4Au/CeO2-
DP catalyst was very active with a conversion reaching 93% atroom temperature.This high activity has been already relatedby numerous authors [23–26] to the presence of nanoparticleson the catalyst surface. All the above results confirm thatthe presence of gold nanoparticles is necessary to oxidizedifferent gaseous molecules at low temperature. In fact, only
4 The Scientific World Journal
Table 1: Characteristic temperatures obtained for the catalyzed propylene, toluene, carbon monoxide, and carbon black oxidation reactions.
small nanoparticles are present in the 4Au/CeO2-DP solid,
while only 20% of the particles was smaller than 10 nmin the 4Au/CeO
2-Imp solid [21, 22]. Moreover, among the
impregnated catalysts, the one containing less gold is moreactive in all reactions. This confirms that the gold particlessize is amajor determining factor in the catalytic performanceof these solids. On the other hand, the high chlorine contentresulting from HAuCl
4precursor in the impregnated solids
may be another reason for its lower performance [2–4].In addition, numerous authors [27, 28] have related thecatalytic activity of Au/CeO
2catalysts to the presence of
positively charged gold species. It was demonstrated using theXPS technique that the quantity of Au+ ions present in the4Au/CeO
2-DP catalyst is two times larger than that present
in the 4Au/CeO2-Imp catalyst [21, 22]. This can also explain
the better performance of the DP solid.
3.2.2. Oxidation of Carbon Black. Figure 2 shows the DSCcurves obtained during carbon black oxidation over thedifferent prepared solids. All TG curves show a 2wt.% weightloss corresponding to the total elimination of CB present inthe mixture (result not shown).
From these latter, two characteristic temperatures can beobtained: 𝑇
𝑖(beginning of CB oxidation) and 𝑇
𝑓(complete
conversion of CB). The maximum of the DSC curve (𝑇max)corresponds to the temperature at which the reaction rate isthe highest. The value Δ𝑇 = (𝑇
𝑓− 𝑇
𝑖) is an indication of the
reaction rate. Table 1 lists the different temperatures relevantto all the reactions considered in this paper.The noncatalyzedoxidation of carbon black was done in the presence of SiC asan inert dilution matrix, and the corresponding DSC curveexhibited one exothermic peak with a maximum at 607∘C.
Mixing tightly CB to calcined ceria leads to a 𝑇max= 368∘C, a gain of 239∘C compared to the noncatalyzedreaction. In fact, due to its oxygen storage capacity (OSC),cerium oxide has been extensively used as a fuel additive,a support for active phases, or even a catalyst in orderto accelerate oxidation reaction. The Ce3+/Ce4+ couple caneasily store and release oxygen, making it mobile and readilyavailable to oxidize different types of adsorbed molecules [16,18, 19]. The addition of gold using the impregnation methodleads to a similar reactivity for the low contents (0.5 wt.% ofAu) but a decreased reactivity for the4Au/CeO
2-Imp catalyst.
This result correlates well with those obtained for gaseousmolecules oxidation.
It seems that the high gold content modified the surfacecharacteristics leading to fewer contact points between CBand labile oxygen sites. For the low gold content, the presence
25 100 175 250 325 400 475 550 625 700D
SC si
gnal
(a.u
.)
CeO2
0.5Au/CeO2-Imp4Au/CeO2-Imp4Au/CeO2-DP
Temperature (∘C)
345∘C
399∘C
368∘C
Figure 2: DSC curves obtained during carbon black (2wt.%) oxida-tion over CeO
2
, 0.5Au/CeO2
-Imp, 4Au/CeO2
-Imp, and 4Au/CeO2
-DP solids.
of nanoparticles along with some large particles did notaffect the catalytic performance. Finally, the presence of4Au/CeO
2-DP catalyst enhanced slightly the catalytic perfor-
mance compared to ceria with a 𝑇max = 345∘C. The mecha-nism by which the presence of gold enhances the oxidationof CB is difficult to determine; however, two scenarios maybe suggested. First, it has been demonstrated that depositedgold is partly present as Au+ species reduce at relativelylow temperatures [21]. This is probably a reason for theenhanced reactivity of the 4Au/CeO
2-DP catalyst. Second, at
moderate temperatures, the vapour pressure of CB increases,and volatile molecules are released. These latter adsorb easilyon gold nanoparticles (similar behaviour as probe moleculesused in Section 3.2.1) and oxidize in the presence of air withthe release of someheat.This phenomenon can be responsiblefor the initiation of CB oxidation at lower temperature inthe presence of 4Au/CeO
2-DP catalyst. Table 1 shows that
the CB oxidation reaction is the fastest in the presence of4Au/CeO
2-Imp solid (ΔT = 72∘C) which is expected as the
reaction takes place at higher temperatures (𝑘 = 𝐴𝑒−𝐸𝑎/𝑅𝑇).However, even if the reaction takes place at lower temperaturein the the presence of the 4Au/CeO
2-DP catalyst, it is faster
(ΔT = 110∘C) compared to the CeO2catalyzed reaction (ΔT
= 136∘C). This result confirms the catalytic activity of the4Au/CeO
2-DP solid. Finally, the catalysts prepared using the
impregnationmethod are less active which is probably due tothe presence of large solid particles and chloride ions.
The Scientific World Journal 5
4. Conclusion
The catalytic performance of xAu/CeO2solids prepared by
the deposition-precipitation and the impregnation methodswas evaluated in different oxidation reactions. The solidprepared by the deposition-precipitation method was themost active in all the oxidation reactions. This was related tothe presence of a higher percentage of reactive Au+ speciesin this solid compared to its percentage in the impregnatedsolids. The large number of nanoparticles in the 4Au/CeO
2-
DP catalyst is also a determining factor in its reactivitytowards propylene, toluene, carbon monoxide, and carbonblack oxidation reactions. It is also shown that the chloridecontent is another factor that negatively affects the catalyticreactivity. These results can be considered as a proof that theDP solids aremore active in the oxidation reactions comparedto the Imp solids.However, according to [22], it seems that theImp solids are more relevant for a different type of reactionsthat will be detailed in a forthcoming work.
References
[1] M. Haruta, T. Kobayashi, H. Sano, and N. Yamada, “Novel goldcatalysts for the oxidation of carbonmonoxide at a temperaturefar below 0∘C,” Chemistry Letters, vol. 16, no. 2, pp. 405–408,1987.
[2] F. Boccuzzi, A. Chiorino, S. Tsubota, and M. Haruta, “An IRstudy of CO-sensing mechanism on Au/ZnO,” Sensors andActuators B, vol. 25, no. 1, pp. 540–543, 1995.
[3] G. C. Bond and D. T. Thompson, “Gold-catalysed oxidation ofcarbon monoxide,” Gold Bulletin, vol. 33, no. 2, pp. 41–50, 2000.
[4] M. Haruta and M. Date, “Advances in the catalysis of Aunanoparticles,” Applied Catalysis A, vol. 222, no. 1-2, pp. 427–437, 2001.
[5] M.Haruta, “Nanoparticulate gold catalysts for low-temperatureCO oxidation,” Journal of New Materials for ElectrochemicalSystems, vol. 7, no. 3, pp. 163–172, 2004.
[6] S. Minico, S. Scire, C. Crisafulli, and S. Galvagno, “Influence ofcatalyst pretreatments on volatile organic compounds oxidationover gold/iron oxide,”Applied Catalysis B, vol. 34, no. 4, pp. 277–285, 2001.
[7] S. Minico, S. Scire, C. Crisafulli, R. Maggiore, and S. Gal-vagno, “Catalytic combustion of volatile organic compounds ongold/iron oxide catalysts,” Applied Catalysis B, vol. 28, no. 2–4,pp. 245–251, 2000.
[8] M. A. Centeno, M. Paulis, M. Montes, and J. A. Odrio-zola, “Catalytic combustion of volatile organic compounds onAu/CeO
2
/Al2
O3
and Au/Al2
O3
catalysts,” Applied Catalysis A,vol. 234, no. 1-2, pp. 65–78, 2002.
[9] S. Scire, S. Minico, C. Crisafulli, C. Satriano, and A. Pis-tone, “Catalytic combustion of volatile organic compounds ongold/cerium oxide catalysts,” Applied Catalysis B, vol. 40, no. 1,pp. 43–49, 2003.
[10] C. Gennequin, M. Lamallem, R. Cousin, S. Siffert, F. Aıssi, andA. Aboukaıs, “Catalytic oxidation of VOCs on Au/Ce-Ti-O,”Catalysis Today, vol. 122, no. 3-4, pp. 301–306, 2007.
[11] M. Lamallem, H. E. Ayadi, C. Gennequin et al., “Effect of thepreparation method on Au/Ce-Ti-O catalysts activity for VOCsoxidation,” Catalysis Today, vol. 137, no. 2-4, pp. 367–372, 2008.
[12] M. Haruta, A. Ueda, S. Tsubota, and R. M. T. Sanchez, “Low-temperature catalytic combustion of methanol and its decom-posed derivatives over supported gold catalysts,” CatalysisToday, vol. 29, no. 1–4, pp. 443–447, 1996.
[13] A. Wolf and F. Schuth, “A systematic study of the synthesisconditions for the preparation of highly active gold catalysts,”Applied Catalysis A, vol. 226, no. 1-2, pp. 1–13, 2002.
[14] M.Hosseini, T. Barakat, R. Cousin et al., “Catalytic performanceof core-shell and alloy Pd-Au nanoparticles for total oxidationof VOC: the effect of metal deposition,”Applied Catalysis B, vol.111-112, pp. 218–224, 2012.
[15] M. Itoh, M. Saito, M. Takehara, K. Motoki, J. Iwamoto, andK. Machida, “Influence of supported-metal characteristics ondeNO
𝑥
catalytic activity over Pt/CeO2
,” Journal of MolecularCatalysis A, vol. 304, no. 1-2, pp. 159–165, 2009.
[16] S. Aouad, E. Abi-Aad, and A. Aboukaıs, “Simultaneous oxi-dation of carbon black and volatile organic compounds overRu/CeO
2
catalysts,”AppliedCatalysis B, vol. 88, no. 3-4, pp. 249–256, 2009.
[17] P. A. Carlsson and M. Skoglundh, “Low-temperature oxidationof carbon monoxide and methane over alumina and ceriasupported platinum catalysts,” Applied Catalysis B, vol. 101, no.3-4, pp. 669–675, 2011.
[18] D. Homsi, S. Aouad, J. El Nakat et al., “Carbon black andpropylene oxidation over Ru/Ce
𝑥
Zr1−𝑥
O2
catalysts,” CatalysisCommunications, vol. 12, no. 8, pp. 776–780, 2011.
[19] S. Aouad, E. Saab, E. Abi-Aad, and A. Aboukaıs, “Study of theRu/Ce system in the oxidation of carbon black and volatileorganic compounds,” Kinetics and Catalysis, vol. 48, no. 6, pp.835–840, 2007.
[20] E. Saab, S. Aouad, E. Abi-Aad, M. N. Bokova, E. A. Zhilinskaya,and A. Aboukaıs, “Carbon black oxidation mechanism in looseand tight contacts with Al
2
O3
and CeO2
catalysts,” Kinetics andCatalysis, vol. 48, no. 6, pp. 841–846, 2007.
[21] A. Aboukaıs, S. Aouad, H. El-Ayadi et al., “Physicochemicalcharacterization of Au/CeO
2
solid. Part 1: the deposition-precipitation preparation method,” Materials Chemistry andPhysics, vol. 137, pp. 34–41, 2012.
[22] A. Aboukaıs, S. Aouad, H. El-Ayadi et al., “Physicochemicalcharacterization of Au/CeO
2
solids. Part 2: the impregnationpreparation method,”Materials Chemistry and Physics, vol. 137,no. 1, pp. 42–47, 2012.
[23] L. H. Chang, N. Sasirekha, B. Raiesh, and Y. W. Chen, “COoxidation on ceria—and manganese oxide-supported gold cat-alysts,” Separation and Purification Technology, vol. 58, no. 1, pp.211–218, 2007.
[24] L. C. Wang, Q. Liu, X. S. Huang, Y. M. Liu, Y. Cao, and K. N.Fan, “Gold nanoparticles supported on manganese oxides forlow-temperature CO oxidation,”Applied Catalysis B, vol. 88, no.1-2, pp. 204–212, 2009.
[25] O. H. Laguna, M. A. Centeno, F. Romero-Sarria, and J. A.Odriozola, “Oxidation of CO over gold supported on Zn-modified ceria catalysts,”Catalysis Today, vol. 172, no. 1, pp. 118–123, 2011.
[26] M. Mendez-Cruz, J. Ramırez-Solıs, and R. Zanella, “CO oxi-dation on gold nanoparticles supported over titanium oxidenanotubes,” Catalysis Today, vol. 166, no. 1, pp. 172–179, 2011.
[27] S. Minico, S. Scire, C. Crisafulli, A. M. Visco, and S. Galvagno,“FT-IR study of Au/Fe
2
O3
catalysts for CO oxidation at lowtemperature,” Catalysis Letters, vol. 47, no. 3-4, pp. 273–276,1997.
6 The Scientific World Journal
[28] E. D. Park and J. S. Lee, “Effects of pretreatment conditions onCO oxidation over supported Au catalysts,” Journal of Catalysis,vol. 186, pp. 1–11, 1999.
